Abstract
BACKGROUND: Anger has been suggested as a risk factor for stroke. Perceived social support (PSS) may relieve anger, thus reducing the risk of stroke; however, evidence supporting this is limited. We aimed to examine whether PSS modifies the risk of stroke associated with anger expression. METHODS: A cohort study was conducted among 1,806 community residents aged 40-74 years who received a cardiovascular risk survey, including anger expression, in 1997. A Cox proportional hazards model was applied to the participants with low and high PSS to calculate the hazard ratios (HRs) and 95% confidence intervals (CIs) of the risks of total stroke and its subtypes based on total anger expression after adjusting for known stroke risk factors. RESULTS: The median follow-up duration was 18.8 years, with 51 incident strokes. Among the participants with low PSS, anger expression had a positive association with the total stroke risk: The multivariable HR per standard deviation increment of total anger expression was 1.43 (95% CI, 1.13-1.82). In contrast, no association was identified among those with high PSS. The corresponding HR was 0.83 (95% CI, 0.49-1.40), with a significant interaction between low and high PSS (P = 0.037). Similar associations regarding the risk of ischemic stroke were found. CONCLUSION: We found an increased risk of stroke associated with anger expression among the participants with low PSS, but not among those with high PSS. Our results suggest that PSS might mitigate the risk of stroke associated with anger.